For agencies
An AI visibility audit your clients get under your name
A branded report for your agency that shows how ChatGPT, Perplexity, Grok and Google AI Mode answer for your client’s category: who is named, what is cited, and the 5 fixes. Delivered in 48 hours.
Clients are asking whether ChatGPT recommends them. I run their buyers’ questions in ChatGPT, Perplexity, Grok and Google AI Mode and hand you a finished report with your logo on it: who is named, what is cited, and the 5 fixes. You send it to your client. I never do.
Who this is for
- You run a marketing, SEO, content or PR agency, and clients now ask whether ChatGPT and Google AI Mode recommend them.
- You want to answer with evidence from the answers themselves, not a tool score you cannot explain.
- You do not have a person free to run buyer questions across four AI engines and write up what they saw.
- You want the report to carry your name and logo, and your relationship with the client to stay yours.
What your agency receives
A document under your agency’s name, with these sections.
- 1. Summary and scoreOne paragraph for the client: in how many of the buyer questions each engine named them, and the single most surprising finding.
- 2. Buyer questions testedUp to 10 questions the client’s buyers ask: best X for Y, X vs a competitor, alternatives to a competitor, how to solve the problem, and the brand name itself.
- 3. One table per engineChatGPT, Perplexity, Grok and Google AI Mode. For every question: is the client named, in what position, which companies are named, and every URL cited.
- 4. Who is named insteadThe competitors that appear most often across all answers, with how many answers each appears in.
- 5. Where the answers come fromThe top cited domains across all engines: listicles, Reddit threads, review sites, LinkedIn posts, publisher pages.
- 6. The 5 fixes, in orderSpecific pages to create, specific cited listicles and threads and how to get into them, and the LinkedIn and Reddit actions that fit. Each fix says which answers it targets.
- 7. Method and capture dateThe engines, the date each answer was captured, and a statement that every finding was seen on screen in that run.
How it works
- 1You send four thingsClient company name, client website, the client’s category in a few words, and your agency logo. You may add up to 10 buyer questions, or leave them to me.
- 2I run the questionsEach question goes into ChatGPT, Perplexity, Grok and Google AI Mode. I record only what appears on screen: who is named, in what order, and what is cited.
- 3You get the report in 48 hoursA document under your agency’s name and logo, ready to send. There is no mention of me in it.
- 4The client stays yoursI never contact your client, never ask for their accounts, and never mention that I did the research.
Why now
Buyers have moved part of their research into AI answers, and fewer of them click through afterwards. Every figure below links to its publisher.
51% of B2B software buyers now start their research with an AI chatbot more often than with Google, up from 29% in 2025. Survey of 1,076 buyers and decision-makers, March 2026. G2, 15 Apr 2026.
8% vs 15% Google users clicked a traditional result link in 8% of visits when an AI summary appeared, against 15% of visits without one. Browsing data from 900 US adults, March 2025. Pew Research Center, 22 Jul 2025.
1% of visits to a Google results page with an AI summary included a click on a link inside the summary (same Pew study). Pew Research Center, 22 Jul 2025.
Starting point Forrester states that genAI searches are the starting point for B2B buyers. The release gives no percentage for this statement. Forrester, 21 Jan 2026.
My own check, 17 Oct 2026: Google AI Mode named me in 1 of 12 buyer queries and cited this site in none. Read the case study.
Questions
Is the report really white-label?
Yes. The document carries your agency’s name in the title and header and your logo. It does not mention me, and I never contact your client.
Which AI engines do you test?
ChatGPT, Perplexity, Grok and Google AI Mode. Each buyer question is run in each engine, and the answer is recorded as it appears on screen on the capture date.
How fast is it?
The report is delivered within 48 hours of receiving the client name, website, category and your logo.
Can the engines give different answers on another day?
Yes. AI answers change between runs and between users. The report states the capture date and describes what appeared in that run, not a permanent ranking.
What does the agency need to provide?
The client’s company name, website and category, and your logo. Up to 10 buyer questions are optional: if you send none, I write them from the client’s website.
Do you guarantee the client will be named after the fixes?
No. The 5 fixes are the changes most likely to get the client cited, based on which sources the engines cited in the audit. No one controls what an AI engine says.
How do I get a sample?
Use the form on this page, or book the 30-minute call. The sample audit covers one of your clients, and you decide whether to show it to them.